Title: Combinatory method of ant colony optimisation: artificial neural network in intelligent control systems for diesel engines to reduce the emissions and improve the performance, using laboratory experiments

Authors: Abbas Zarenezhad Ashkezari; Elham Askari

Addresses: Department of Mechanical Engineering, Imam Khomeini Marine Sciences University, Nowshahr, 4651783311, Iran ' Department of Computer Engineering, Fouman and Shaft Branch, Islamic Azad University, Fouman, 4351835875, Iran

Abstract: In this work, by using ant colony optimisation (ACO) algorithm, analysing and optimisation of the NOx emissions, and fuel consumption in a diesel engine are done by applying controllable variables of engine speed, inlet air temperature, and fuel mass rate. For this purpose, by using of experimental tests, the necessary requirements for modelling of the input variables and the output parameters were provided via artificial neural network (ANN), and the ACO algorithm was applied to reduce NOx and bsfc simultaneously. The results showed that, the application of ACO algorithm to the modelling led to subsequent 28% and 5% decrease in NOx and bsfc, respectively. Moreover, due to rapid convergence and significant optimisation of the output parameters, the combinatory method of ACO - ANN can be used as an effective method in intelligent control systems for diesel engines in order to reduce emissions as well as fuel consumption.

Keywords: ANN; artificial neural network; ant colony; NOx; bsfc; diesel engines.

DOI: 10.1504/IJHVS.2022.127027

International Journal of Heavy Vehicle Systems, 2022 Vol.29 No.3, pp.305 - 325

Received: 15 Jul 2021
Accepted: 10 Oct 2021

Published online: 18 Nov 2022 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article